2026-03-09T05:18:57Z https://ipsj.ixsq.nii.ac.jp/oai

oai:ipsj.ixsq.nii.ac.jp:00055064 2025-01-22T05:36:37Z 1164:4961:4979:4982

異本解析を目的としたオリジナル文書抽出モデルの考察 Model to Extract Original Sources from the Variant Documents 三宅, 真紀 赤間, 啓之 馬越庸恭 中川, 正宣 Maki, Miyake ，Hiroyuki, Akama Nobuyasu, Makoshi Masanori, Nakagawa 本研究では、異本文書から共通したオリジナル文書を抽出するモデルを考案し、新約聖書学の「共観福音書問題」に適用を試みた。そして、人工的に作成した文書（オリジナル文書）を用いて、異本文書からオリジナルの文書の抽出を確認し、提案したモデルの妥当性について検証した。 In this paper, we propose a quantitative model to extract original sources from the variant documents. The model is evaluated by simulating a process of intertextuality between the supposed original texts and their derived and modified versions. The first step is to create some artificial texts, which we assume to be dummy "original texts"(in abbreviation, Os), and some "variant documents"(Vs) by randomly distributing all the words contained in each of the Os. Secondly we try to see whether our proposed model is able to extract the traces of the Os in the form of factors, by applying factor analysis to the lexical frequency data gathered from all the subsets of the joint Vs. Based on the result of the simulation, we apply this model to the synoptic problem, which is about the genealogical interdependence between the Synoptic Gospels as one of the controversial subjects in the New Testament (NT) studies. technical report 情報処理学会 2005-05-27 application/pdf 情報処理学会研究報告人文科学とコンピュータ（CH） 51(2005-CH-066) 2005 39 46 AN1010060X https://ipsj.ixsq.nii.ac.jp/record/55064/files/IPSJ-CH05066006.pdf jpn